On Tue, Sep 15, 2009 at 10:25 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> On Tue, Sep 15, 2009 at 10:10 PM, Robert Haas <robertmhaas@gmail.com> wrote:
>>> * I'm not sure about this, because surely you would have tested it,
>>> but isn't it looking at the wrong side of the join clauses? I thought
>>> the idea is to prove the nullable (inner) side of the join unique.
>>
>> Grr. I think it's more broken than that. Wow, this is really embarassing.
>
> Well, you're definitely right that it's looking at the wrong side of
> the join clauses. Still trying to figure out if there is another bug,
> too.
It looks to me like relation_is_distinct_for() is also horribly broken
in my previous version. I think the attached is how it is supposed to
work.
...Robert